A miniaturized MR1 metabolite display system with native-like protein features

Photis Rotsides
Omkar Shinde
Julia N Danon
Nikolaos G Sgourakis author has email address

Center for Computational and Genomic Medicine, Department of Pathology and Laboratory Medicine, The Children’s Hospital of Philadelphia, Philadelphia, United States
Department of Biochemistry and Biophysics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, United States

https://doi.org/10.7554/eLife.111511.1

Open access
Copyright information

Figures and data

Design and purification of SMART-MR1.
A. Crystal structure of human MR1/β₂m (PDB 6PUC) superimposed onto the crystal structure of human HLA-A*02:01/β₂m (PDB 9SKO). Insets illustrate key interactions that stabilize the β₂m interface with the α₁/α₂ domains of HLA-A*02:01/MR1. Polar and hydrophobic contacts are indicated by yellow and black dashes, respectively. B. AlphaFold model of SMART-MR1 using the α₁/α₂ domain of MR1 and the stabilizing domain of SMART MHC. The inset displays contacts between the stabilizing domain and the α₁/α₂ domain of MR1. The rigid linker connecting the two domains is colored black. C. Size exclusion chromatography (SEC) trace of SMART-MR1 refolded in the presence of 6-FP. D. Melting temperatures (T_m, °C) obtained from DSF of SMART-MR1 and hpMR1/bβ₂m refolded in the presence of 6-FP. Data are mean ± s.d. for n = 3 technical replicates. Unpaired t-test was performed to obtain a P value of 0.6713, P > 0.05 (ns).

Ligand dependence of in vitro refolding and thermal stability of SMART-MR1 complexes.
A. Chemical structures of ligands tested. B. Refolding yields (total mg protein) from SEC purification of SMART-MR1 refolded in the presence of ligands. C. T_m values obtained from DSF experiments of SMART-MR1 refolded in the presence of ligands. Data are mean ± s.d. for n = 3 technical replicates. D. Correlation plot of T_m versus refolding yield with the coefficient of determination from linear regression (R²) shown.

Solution biophysics of SMART-MR1.
A. 2D ¹H-¹⁵N TROSY of 650 μM SMART-MR1/6-FP recorded at a ¹H field of 800 MHz at 25 °C. B. Competitive binding of TAMRA-labelled JYM20 to SMART-MR1/DCF as a function of increasing Ac-6-FP concentration, measured by fluorescence polarization. Data are mean ± s.d. for n = 3 technical replicates. One-way ANOVA was performed relative to conditions without Ac-6-FP, P < 0.0002(***), and P < 0.0001(****). C. ITC data titrating 200 μM A-F7 TCR into a sample containing 20 μM SMART-MR1/5-OP-RU. Black line is the fit of the isotherm with the K_D and thermodynamic values determined using a 1-site binding model (n = 1).

Cryo-EM structure of SMART-MR1/5-OP-RU bound to A-F7 TCR.
A. Cryo-EM structure of SMART-MR1/5-OP-RU superimposed onto the AlphaFold model of SMART-MR1 and the crystal structure of human MR1/hβ₂m-5-OP-RU (PDB 6PUC). The inset shows the sidechain of W17 of SMART-MR1 and W60 of hβ₂m. B. Conformation of 5-OP-RU in the SMART-MR1 Cryo-EM structure superimposed onto 5-OP-RU from the native MR1 crystal structure. C. Close up of 5-OP-RU within the ligand binding groove of native and SMART-MR1. Polar and hydrophobic contacts are indicated by yellow and black dashes, respectively. Residues are colored as in panel A. D. Cryo-EM structure of A-F7 TCR bound to the α₁/α₂ domain of SMART-MR1/5-OP-RU (cyan) superimposed onto the native MR1/5-OP-RU-A-F7 crystal structure (grey). The inset highlights key interactions between the TCR and the ligand as well as residues within the MR1 groove. E. Top view of CDR loop interactions with the surface of SMART-MR1/5-OP-RU. CDR loops are shown as tubes and are colored as in panel D.

Thermal stability of SMART-MR1 compared to full-length MR1.
A. SDS-PAGE gel of purified SMART-MR1 refolded with 6-FP. B. Normalized DSF traces of purified SMART-MR1 and human platform MR1 (hpMR1) refolded with 6-FP. Data are mean ± s.d. for n = 3 technical replicates.

SEC traces and DSF curves of refolded SMART-MR1 complexes.
A. Size exclusion chromatography (SEC) traces of SMART-MR1 refolded in the absence (empty) and presence of ligand. SEC traces for MR1 are color coded as shown in panel C. B. Normalized DSF traces of purified empty and ligand loaded SMART MR1. C. Summary of refolding yields obtained from SEC experiments and melting temperatures (T_m) obtained from DSF experiments. Data are mean ± s.d. for n = 3 technical replicates.

Cryo-EM data processing for A-F7 TCR in complex with SMART-MR1-5-OP-RU.
A. Representative micrograph of the collection for A-F7 TCR/SMART-MR1-5-OP-RU complex. Scale bar in yellow is 200 Å. B. Representative 2D classes of A-F7 TCR/SMART-MR1-5-OP-RU complex. Scale bar in white is 150 Å. C. Cryo-EM data processing workflow of A-F7 TCR/SMART-MR1-5-OP-RU complex. D. Local-resolution estimation of reconstructed map as determined within CryoSPARC. E. Gold-standard FSC curves used for global-resolution estimates within CryoSPARC. F. Viewing direction distribution of the reconstructed map.

Quality of SMART-MR1-A-F7 cryo-EM map.
A. Cryo-EM density of 5-OP-RU and the residues interacting with it. MHC is colored cyan, TCRα in blue, TCRβ in pink, and 5-OP-RU in light brown. Map is contoured at 0.196. B. Cryo-EM densities of CDRs of TCRα. Maps are contoured at 0.150 (CDR1α), 0.200 (CDR2α), and 0.180 (CDR3α). C. Cryo-EM densities of CDRs of TCRβ. Maps are contoured at 0.240 (CDR1β), 0.140 (CDR2β), and 0.120 (CDR3β). D-F. Cryo-EM densities of stabilization domain helices. Maps are contoured at 0.095 (D), 0.085 (E), and 0.070 (F).

Cryo-EM data collection, refinement and validation statistics SMART-MR1/A-F7 TCR

Cryo-EM data collection, refinement and validation statistics SMART-MR1/A-F7 TCR

Sign up for email alerts